What is C2PA?
The Coalition for Content Provenance and Authenticity (C2PA) is an open technical standard developed by companies including Adobe, Microsoft, Truepic, and Intel. It defines how digital photos can carry verifiable cryptographic provenance metadata detailing who created an asset, what editing tools were used, and whether artificial intelligence was involved.
How C2PA is stored in image files
C2PA data is encapsulated using ISO/IEC 19566-5 JPEG Universal Metadata Box Format (JUMBF).
- In JPEG: JUMBF boxes are stored inside one or more
APP11(0xFFEB) segments near the start of the file. - In PNG: Content Credentials are stored in a dedicated
caBXancillary chunk. - In WebP: C2PA boxes are stored in RIFF chunks.
Why authentic camera photos receive AI badges
When you retouch a photograph in Adobe Photoshop (for example, using Generative Fill to remove a background distraction) or use Adobe Lightroom's Generative Remove feature, Adobe automatically attaches a signed C2PA manifest. Ingestion scanners on Instagram, Facebook, and TikTok check for these manifests; upon finding one, their algorithms flag the entire original photograph with an "AI info" label.
How to inspect and remove C2PA losslessly
Clean inspects the APP11 segment or caBX chunk and parses the claim generator, signature issuer, and assertions. If you choose to clean the file, Clean removes the C2PA container directly from the file header while copying your compressed image bytes byte-for-byte. Your pixels stay bit-identical.